Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    (tomorrow)
    Lance Parsons
    @lparsons
    ;-)
    Thanks again, you've gone above and beyond (as usual).
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    no worries
    Martin Cech
    @martenson
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    anyone opposed to a new ephemeris release ?
    0.9.0 is seriously outdated
    Björn Grüning
    @bgruening
    :+1:
    John Chilton
    @jmchilton
    it is desperately needed
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    hmm, how is make add-history supposed to work ?
    John Chilton
    @jmchilton
    make add-history ITEM=pr185
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    ah, alright
    John Chilton
    @jmchilton
    item can also be a commit hash or I think issue345
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    neat, that works
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    gotta run, but if anyone wants to finish feel free to push to galaxyproject/ephemeris#132
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    Can I get a review on galaxyproject/ephemeris#134 ? I'll do a patch release for this one
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    alright, thanks @bgruening. bugfix release 0.10.1 is out
    Björn Grüning
    @bgruening
    Cool! Thanks for the fix.
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    hmm, anyone knows who owns the ephemeris readthedocs ? it hasn't been updated since 0.9.1
    Martin Cech
    @martenson
    I suspect @jmchilton
    John Chilton
    @jmchilton
    I’ll update them
    Lance Parsons
    @lparsons
    Hi all. Not much traffic here, but I thought I'd start here. I'm testing out using Ephemeris for tool management. I run galaxy behind an apache proxy. Trying to install tools seems like it might be installing the tools OK, but ephemeris returns an error "502 Proxy Error". Seems like this was discussed in galaxyproject/ansible-galaxy-tools#63 (@mvdbeek, @bernt-matthias). Is there something I can do to avoid the error, or would a patch to ephemeris help? I'm happy to help sort it out, I'd really like to get using ephemeris for tool mgmt. Thx.
    M Bernt
    @bernt-matthias
    I never found a way to fix this. Only nginx (seemingly) helped in the end.
    Lance Parsons
    @lparsons
    nginx is not an option for me right now, I need the Apache CAS auth plugin...
    Any ideas on what is going on and where I might be able to tweak things? Get ephemeris to try again perhaps?
    M Bernt
    @bernt-matthias
    As far as I got it the triggered installation of the tools just needs quite long .. which leads to the time out. If I remember correctly it was discussed if the tool installations should simply be put in some work queue. This would require changes on the Galaxy side. Questing is also how to poll the status of the installation...
    As a hack one might also try to run ephemeris(/ansible) directly on the host running Galaxy and shortcut the proxy...?
    Lance Parsons
    @lparsons
    Thanks so much for the info. Not sure what you meant by "Questing is also how to poll the status of the installation...".
    Good idea of running on the host (though not really ideal)
    M Bernt
    @bernt-matthias
    If ephemeris just triggers to put the installation task into some queue .. then one can only know if the queuing was successful. So one needs to implement some means to query the Galaxy server if the tool was installed sucessfully (or if it failed, did not yet start, or installation is still going on)
    Lance Parsons
    @lparsons
    Ah, got it. You meant "queuing" not "questing". ;-)
    Makes total sense.
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    @bernt-matthias @lparsons I think that’s been fixed by just not relying on a 504. What version of ephemeris are you using ? Can you post the error you’re seeing ? It might be as simple as adding a check for 502 to https://github.com/galaxyproject/ephemeris/blob/0b9877319cb2f267ebe2ac577634adfef3390963/src/ephemeris/shed_tools.py#L393
    Derek Benson
    @dnbenso
    Hi all. Is there a URL option for unix sockets? Not a show stopper - just wondering.
    FBernal-oPs
    @FBernal-oPs
    Hi!, I am getting the following error when attempting to install
     shed-tools install -g http://my_galaxy_instance.com --name immuneml_tools --owner knutwa2 -a xxxxxxx -v --toolshed https://galaxy-ntnu.bioinfo.no/toolshed_nels/ --section immuneml_tools
    Storing log file in: /tmp/ephemeris_mo3cv45g
    (1/1) Installing repository immuneml_tools from knutwa2 to section "immuneml_tools" at revision xxxxxx (TRT: 0:00:00.539130)
        * Error installing a repository (after 0:00:00.197149 seconds)! Name: immuneml_tools,owner: knutwa2, revision: xxxxxx, error: {"err_msg": "Uncaught exception in exposed API method:", "err_code": 0}
    Installed repositories (0): []
    Skipped repositories (0): []
    Errored repositories (1): [('immuneml_tools', 'xxxxxx')]
    All repositories have been installed.
    Total run time: 0:00:00.736697
    Björn Grüning
    @bgruening
    revision xxxxxx does not exist :)
    you should enter a real revision number :)
    FBernal-oPs
    @FBernal-oPs
    ephemeris should be run where the galaxy server is, right?
    is there any way to do this remotely?
    ansible role that can do this?
    FBernal-oPs
    @FBernal-oPs
    $ shed-tools install -g http://internal.podman-rootless-t.uiocloud.no --name immuneml_tools --owner knutwa2 -a b9450e14a90b3c7eda3acf0cb9a4bf1d -v --toolshed https://galaxy-ntnu.bioinfo.no/toolshed_nels/ --section immuneml
    Storing log file in: /tmp/ephemeris_mjk4ozdm
    (1/1) Installing repository immuneml_tools from knutwa2 to section "immuneml" at revision 6a4ade98ee47 (TRT: 0:00:00.089474)
        * Error installing a repository (after 0:00:00.169407 seconds)! Name: immuneml_tools,owner: knutwa2, revision: 6a4ade98ee47, error: {"err_msg": "Uncaught exception in exposed API method:", "err_code": 0}
    Installed repositories (0): []
    Skipped repositories (0): []
    Errored repositories (1): [('immuneml_tools', '6a4ade98ee47')]
    All repositories have been installed.
    Total run time: 0:00:00.259305
    Björn Grüning
    @bgruening
    You can do this from any computer as long as you can reach Galaxy from that host.
    I would recommend to use an IUC tool to test
    not sure if this tool is a recent tool, updated and has conda dependencies
    Ghost
    @ghost~5772e7e2c2f0db084a206e1b
    @FBernal-oPs you will need to look at the Galaxy logs, something did not go right on the Galaxy end
    alesss78
    @alesss78
    @alesss78
    hi all, I have a problem with a MacOS installed version of Galaxy: under the "Shared data" I don't have access to any public resource (Data Libraries, histories etcetera). Should I configure my Galaxy installation in order to have access to the same public resources that I find on http://usegalaxy.eu?
    And if yes, I would need help configuring it
    I don't know if this is the correct room, If out of scope please forgive and pass on
    M Bernt
    @bernt-matthias
    @alesss78 I guess this room is better: https://gitter.im/galaxyproject/admins
    kenza12
    @kenza12

    Hello, I want to install a new tool that I added in https://testtoolshed.g2.bx.psu.edu ,to my local galaxy. I just discovered Ephemeris. I tried this command: shed-tools install --name retropath2 --owner tduigou --toolshed "https://testtoolshed.g2.bx.psu.edu" --section_label "Test_RETROPATH2" -g "http://localhost:8080" -a "API_KEY" -v
    I got this error:

    Storing log file in: /tmp/ephemeris_57yxtfkq
    (1/1) Installing repository retropath2 from tduigou to section "Test_RETROPATH2" at revision 0f2084b71072 (TRT: 0:00:00.714898)
        * Error installing a repository (after 0:00:00.879750 seconds)! Name: retropath2,owner: tduigou, revision: 0f2084b71072, error: {"err_msg": "Uncaught exception in exposed API method:", "err_code": 0}
    Installed repositories (0): []
    Skipped repositories (0): []
    Errored repositories (1): [('retropath2', '0f2084b71072')]
    All repositories have been installed.
    Total run time: 0:00:01.595166

    and in galaxy logs:

    urllib3.connectionpool DEBUG 2021-06-30 13:57:11,089 [p:68,w:1,m:0] [uWSGIWorker1Core2] Starting new HTTPS connection (1): testtoolshed.g2.bx.psu.edu:443
    urllib3.connectionpool DEBUG 2021-06-30 13:57:11,953 [p:68,w:1,m:0] [uWSGIWorker1Core2] https://testtoolshed.g2.bx.psu.edu:443 "GET /api/repositories/get_repository_revision_install_info?name=retropath2&owner=tduigou&changeset_revision=0f2084b71072 HTTP/1.1" 200 None
    galaxy.tool_shed.galaxy_install.repository_dependencies.repository_dependency_manager DEBUG 2021-06-30 13:57:11,954 [p:68,w:1,m:0] [uWSGIWorker1Core2] Creating repository dependency objects...
    galaxy.web.framework.decorators ERROR 2021-06-30 13:57:11,955 [p:68,w:1,m:0] [uWSGIWorker1Core2] Uncaught exception in exposed API method:
    Traceback (most recent call last):
      File "lib/galaxy/web/framework/decorators.py", line 294, in decorator
        rval = func(self, trans, *args, **kwargs)
      File "lib/galaxy/webapps/galaxy/api/tool_shed_repositories.py", line 144, in install_repository_revision
        payload)
      File "lib/galaxy/tool_shed/galaxy_install/install_manager.py", line 713, in install
        install_options
      File "lib/galaxy/tool_shed/galaxy_install/install_manager.py", line 763, in __initiate_and_install_repositories
        self.handle_tool_shed_repositories(installation_dict)
      File "lib/galaxy/tool_shed/galaxy_install/install_manager.py", line 629, in handle_tool_shed_repositories
        new_tool_panel_section_label=new_tool_panel_section_label)
      File "lib/galaxy/tool_shed/galaxy_install/repository_dependencies/repository_dependency_manager.py", line 154, in create_repository_dependency_objects
        all_required_repo_info_dict = self.get_required_repo_info_dicts(tool_shed_url, repo_info_dicts)
      File "lib/galaxy/tool_shed/galaxy_install/repository_dependencies/repository_dependency_manager.py", line 398, in get_required_repo_info_dicts
        url = build_url(tool_shed_url, pathspec=pathspec)
      File "lib/galaxy/util/__init__.py", line 1662, in build_url
        assert parsed_url.scheme in ('http', 'https', 'ftp'), 'Invalid URL scheme: %s' % scheme
    AssertionError: Invalid URL scheme: http
    127.0.0.1 - - [30/Jun/2021:13:57:11 +0000] "POST /api/tool_shed_repositories/new/install_repository_revision HTTP/1.1" 500 - "-" "python-requests/2.25.1"
    [pid: 68|app: 0|req: 7781/7781] 127.0.0.1 () {38 vars in 591 bytes} [Wed Jun 30 13:57:11 2021] POST /api/tool_shed_repositories/new/install_repository_revision => generated 71 bytes in 877 msecs (HTTP/1.1 500) 3 headers in 158 bytes (1 switches on core 2)

    Then, I tried to install another tool from the main toolshed, it works and it could install BWA:
    shed-tools install --name bwa --owner devteam --toolshed "https://toolshed.g2.bx.psu.edu" --section_label "Test_BWA" -g "http://localhost:8080" -a "API_KEY" -v
    Thanks for you help and your time !

    kenza12
    @kenza12
    I think I found the problem. Galaxy Test Tool Shed is not installed on my local galaxy that's why it works only on the main toolshed.